Contributing data

This section is a step-by-step guide for preparing and formatting a new dataset so it can be incorporated into Brasil Data Commons. It’s the import side of the platform — distinct from the API guides, which are about reading data that’s already loaded.

What you’ll produce

To bring a dataset into Data Commons you assemble, at most, three kinds of file:

• CSV file(s) holding the observations themselves.
• MCF file(s) — only when you need to define new statistical variables. If every variable you use already exists, you can skip these entirely.
• A JSON config file that maps your data onto concepts already in Data Commons.

The workflow

1. Document your source — fill in a metadata template describing the original dataset and where it came from.
2. Format the values — shape your observations into the standard entity, date, variable, value CSV.
3. Define new variables — if your data needs statistical variables that don’t exist yet, declare them in an MCF file.
4. Write the config — describe your input files, their sources, and any custom column names in config.json.
5. Validate and deliver — lint the files, record the transformations you applied, and package everything for submission.

Note

This guide is deliberately practical. It focuses on the conventions and standards adopted by Brasil Data Commons, some of which are specific to Brazil and may differ from other Data Commons instances. For the deeper conceptual background, see the official Data Commons guide, Prepare and load your own data.